> Andrew asks for "how to deal with information that falls outside of"
> a scientific paradigm. The question is more general - ie. How to
> deal with information that falls outside of your existing paradigms
> whatever they are?
> To offer a beginning in that matter.
> At first, information that falls outside of one's own existing
> paradigms isn't information. It isn't even data. It's noise.
Kuhn calls this noise "anomalies".
> So the first thing is to pay attention to noise. To monitor it. To
> begin to get curious, interested, engaged with what occurs as noise.
> This is not an easy task. The term means that there is no meaning
> within one's existing structures of meaning.
> Once one is curious, then noise becomes data. What then opens up is
> the possibility of developing theories or hypotheses to experiment
> with. In other words, one creates design principles which will
> produce patterns (or fail to) from the data and the process of
> mapping new territory has begun.
Michael, let me step back a bit and articulate some of the
assumptions behind these paragraphs (I think we're on the same
planet ...).
To me, paradigms are a tool for structuring thinking about the
world, and this is articulated through a domain of discourse (or a
set of concept definitions and relationships). Thomas Kuhn's work is
a major insight to me because he states that paradigms are
initially limited in scope, and are extended over time by a
community of practitioners.
Now, what do we do with this "noise" - how do we learn outside our
current paradigm?. We can start by developing theories and
hypotheses within our current paradigm - this means reusing our
current domain of discourse, or our concept definitions and
relationships, and possibly adding new concepts and relationships. A
paradigm shift occurs, I believe, when our domain of discourse (or
our way of thinking about the world) is inadequate for explaining
some "noise". In this instance, we must challenge some of our
underlying assumptions behind our paradigm, find a new way of
thinking about the world, find a new domain of discourse (often
involving redefinition of existing concepts) and reformulate our
relationships between concepts.
How de we do this? I think if we knew this we'd understand the
creative process (and thus know whether or not it is possible to
build a truly artificially intelligent creature :->). Kuhn describes
the process of somebody recognising the noise, causing a crisis
(which involves testing hypotheses and theories within a paradigm),
which may lead to scientific revolution (when the assumptions on
which the previous paradigm were based are challenged and the world
is redefined).
Is this the process you were describing Michael, and the one you
were interested in Andrew? OR can we describe it better somehow?
-- John O'Neill DSTO C3 Research Centre, Australia email: jao@itd.dsto.gov.au